4 ms·
Interesting question! From some anecdotal experience, large language models struggle with spatial structure (which makes sense given their modality and trainin
by thegeomaster 3y ago
Interesting question!
From some anecdotal experience, large language models struggle with spatial structure (which makes sense given their modality and training data). On the other hand, diffusion models create great images, but this does not translate very well to vector data.
Animation is not a very well researched modality for AI, so it could go either way. It's definitely an interesting direction to consider, as it can democratize motion design even further.
- danielvaughn 3y agoTrue, but in my (limited) understanding, a lottie animation is just json right? I've gotten some really impressive json results with GPT.
- duskwuff 3y agoIt's a JSON structure full of numbers which describe colors and spatial relationships. LLMs have no problem with the JSON syntax, but the numbers they put in it will be nonsense.